Concept graph
Generative AI and LLM glossary
Clear AI definitions with practical context, examples, and links to related models, tools, and concepts.
reinforcement-learning
safe-exploration
A strategy in reinforcement learning that aims to balance exploration and exploitation while minimizing risks.
Models
Sarvam 105B
Sarvam 105B is Sarvam AI's flagship open-weight 105B+ MoE reasoning model for Indian-language chat, coding, long-context work, and agents.
AI platforms
Sarvam AI
Sarvam AI is an India-based sovereign AI platform focused on Indian-language LLMs, speech, translation, document digitization, and enterprise AI agents.
Machine Learning
scalable-dot-product-attention
An efficient variant of attention mechanism designed for large datasets.
Neural Networks
Self-Organizing Map
A type of neural network used for unsupervised learning that maps high-dimensional data to lower dimensions.
Machine Learning
Self-supervised Learning
A type of machine learning where the model learns from unlabeled data by generating its own labels.
Retrieval
Semantic search
Semantic search ranks results by meaning, usually with embeddings, rather than by exact keyword overlap alone.
Computer Vision
Semantic Segmentation
The task of classifying each pixel in an image into a category.
Web Technologies
semantic web
The semantic web is an extension of the World Wide Web that enables data to be shared and reused across applications.
Machine Learning
semi-supervised-learning
A learning approach that combines labeled and unlabeled data for training models.
Natural Language Processing
sentiment-analysis
The use of natural language processing to determine the emotional tone behind words.
Optimization Techniques
Simulated Annealing
A probabilistic technique for approximating the global optimum of a given function.
modeling
simulation
The imitation of the operation of a real-world process or system over time.
Blockchain
Smart Contracts
Self-executing contracts with the terms of the agreement directly written into code.
data analysis
social-network-analysis
The study of social relationships and structures through networks and graph theory.
Product
softmax
Softmax converts a vector of logits into a probability distribution over classes or tokens.
Data Representation
sparse-coding
A representation method that encodes data with a small number of active features.
Signal Processing
sparse-representation
A method in signal processing where signals are represented with a small number of non-zero coefficients.
data-analysis
spatial-temporal-analysis
The study of data that varies across both space and time to uncover patterns and trends.
Inference
Speculative decoding
Speculative decoding speeds generation by drafting tokens with a cheaper model and verifying them with the target model in parallel.
Models
Speech-to-speech
Speech-to-speech, or full-duplex voice, takes spoken audio and returns spoken audio in one live session instead of chaining STT, an LLM, and TTS.
Models
Speech-to-text
Speech-to-text, or ASR, converts audio into text, optionally with timestamps, speakers, and punctuation.
Data Sampling
stratified-sampling
A sampling method that ensures representation from different subgroups in a dataset.
Development
Structured Outputs
Structured Outputs constrain model responses to match a developer-supplied schema, typically for reliable JSON or function-call arguments.